Run splicing cluster assignment, splicing burden, gsva, CLK1

Purpose:

Runs survival analysis models using splicing cluster assignment and 1) single exon splicing burden index (SBI) 2) KEGG Spliceosome GSVA scores or 3) CLK1 exon 4 TPM as a predictor

Usage

Uses a wrapper function (survival_analysis) from utils folder.

Setup

Packages and functions

Load packages, set directory paths and call setup script

library(tidyverse)

── Attaching core tidyverse packages ────────────────────────────────────────────────────────────────── tidyverse 2.0.0 ──
✔ dplyr     1.1.4     ✔ readr     2.1.5
✔ forcats   1.0.0     ✔ stringr   1.5.1
✔ ggplot2   3.5.1     ✔ tibble    3.2.1
✔ lubridate 1.9.4     ✔ tidyr     1.3.1
✔ purrr     1.0.4     
── Conflicts ──────────────────────────────────────────────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag()    masks stats::lag()
ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors

library(survival)
library(ggpubr)
library(ggplot2)
library(patchwork)

root_dir <- rprojroot::find_root(rprojroot::has_dir(".git"))

data_dir <- file.path(root_dir, "data")
analysis_dir <- file.path(root_dir, "analyses", "survival")
input_dir <- file.path(analysis_dir, "results")
results_dir <- file.path(analysis_dir, "results")
plot_dir <- file.path(analysis_dir, "plots")

# If the input and results directories do not exist, create it
if (!dir.exists(results_dir)) {
  dir.create(results_dir, recursive = TRUE)
}

source(file.path(analysis_dir, "util", "survival_models.R"))

Attaching package: 'survminer'

The following object is masked from 'package:survival':

    myeloma

Set metadata and cluster assignment file paths

metadata_file <- file.path(input_dir, "splicing_indices_with_survival.tsv")

cluster_file <- file.path(root_dir, "analyses",
                          "sample-psi-clustering", "results",
                          "sample-cluster-metadata-top-5000-events-stranded.tsv")

kegg_scores_stranded_file <- file.path(root_dir, "analyses",
                          "sample-psi-clustering", "results",
                          "gsva_output_stranded.tsv")

tpm_file <- file.path(data_dir, "rna-isoform-expression-rsem-tpm.rds")
clk1_psi_file <- file.path(root_dir, 
                           "analyses", 
                           "CLK1-splicing_correlations", 
                           "results", 
                           "clk1-exon4-psi.tsv")

Wrangle data Add cluster assignment and spliceosome gsva scores to metadata and define column lgg_group (LGG or non_LGG)

metadata <- read_tsv(metadata_file)

Rows: 707 Columns: 23
── Column specification ──────────────────────────────────────────────────────────────────────────────────────────────────
Delimiter: "\t"
chr  (9): Kids_First_Biospecimen_ID, Histology, Kids_First_Participant_ID, m...
dbl (14): Total, AS_neg, AS_pos, AS_total, SI_A3SS, SI_A5SS, SI_RI, SI_SE, E...

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

clusters <- read_tsv(cluster_file) %>%
  dplyr::rename(Kids_First_Biospecimen_ID = sample_id)

Rows: 752 Columns: 8
── Column specification ──────────────────────────────────────────────────────────────────────────────────────────────────
Delimiter: "\t"
chr (6): sample_id, plot_group, plot_group_hex, RNA_library, molecular_subty...
dbl (2): cluster, group_n

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

clk1_psi <- read_tsv(clk1_psi_file) %>%
  dplyr::rename(CLK1_ex4_PSI = PSI) %>%
  select(-plot_group)

Rows: 752 Columns: 3
── Column specification ──────────────────────────────────────────────────────────────────────────────────────────────────
Delimiter: "\t"
chr (2): Kids_First_Biospecimen_ID, plot_group
dbl (1): PSI

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

gsva_scores <- read_tsv(kegg_scores_stranded_file) %>%
  dplyr::filter(geneset == "KEGG_SPLICEOSOME") %>%
  dplyr::rename(spliceosome_gsva_score = score)

Rows: 23312 Columns: 3
── Column specification ──────────────────────────────────────────────────────────────────────────────────────────────────
Delimiter: "\t"
chr (2): sample_id, geneset
dbl (1): score

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

all_clk4_transcr_counts <- readRDS(tpm_file) %>%
  filter(grepl("^CLK1", gene_symbol)) %>%
  mutate(
    transcript_id = case_when(
      transcript_id %in% c("ENST00000321356.9", "ENST00000434813.3", "ENST00000409403.6") ~ "Exon 4",
#      transcript_id == "ENST00000321356.9" ~ "Exon 4",
      TRUE ~ "Other"
    )
  ) %>%
  group_by(transcript_id) %>%
  summarise(across(starts_with("BS"), sum, na.rm = TRUE)) %>%
  pivot_longer(cols = -transcript_id, names_to = "Kids_First_Biospecimen_ID", values_to = "CLK1_Ex4_TPM") %>%
  filter(transcript_id == "Exon 4") %>%
  inner_join(clusters, by = "Kids_First_Biospecimen_ID") %>%
  left_join(clk1_psi)

Warning: There was 1 warning in `summarise()`.
ℹ In argument: `across(starts_with("BS"), sum, na.rm = TRUE)`.
ℹ In group 1: `transcript_id = "Exon 4"`.
Caused by warning:
! The `...` argument of `across()` is deprecated as of dplyr 1.1.0.
Supply arguments directly to `.fns` through an anonymous function instead.

  # Previously
  across(a:b, mean, na.rm = TRUE)

  # Now
  across(a:b, \(x) mean(x, na.rm = TRUE))

Joining with `by = join_by(Kids_First_Biospecimen_ID)`

# how many clusters?
n_clust <- length(unique(clusters$cluster))

metadata <- metadata %>%
  right_join(all_clk4_transcr_counts %>% dplyr::select(Kids_First_Biospecimen_ID,
                                       cluster, CLK1_Ex4_TPM, CLK1_ex4_PSI)) %>%
  left_join(gsva_scores %>% dplyr::select(sample_id,
                                          spliceosome_gsva_score),
            by = c("Kids_First_Biospecimen_ID" = "sample_id")) %>% 
  dplyr::mutate(cluster = glue::glue("Cluster {cluster}")) %>%
  dplyr::mutate(cluster = fct_relevel(cluster,
                                               paste0("Cluster ", 1:n_clust))) %>%
  dplyr::mutate(lgg_group = case_when(
    Histology == "Low-grade glioma" ~ "LGG",
    TRUE ~ "non-LGG"
  )) %>%
  dplyr::mutate(SBI_SE = SI_SE * 10) %>%
  dplyr::mutate(age_at_diagnosis_years = age_at_diagnosis_days/365.25)

Joining with `by = join_by(Kids_First_Biospecimen_ID)`

Generate coxph models including extent of tumor resection, lgg group, cluster assignment, SBI, and CLK1 exon 4 TPM as covariates

add_model_os <- fit_save_model(metadata[!metadata$extent_of_tumor_resection %in% c("Not Reported", "Unavailable"),],
                              terms = "extent_of_tumor_resection+lgg_group+cluster+age_at_diagnosis_years+SBI_SE+CLK1_Ex4_TPM",
                               file.path(results_dir, "cox_OS_additive_terms_resection_lgg_group_cluster_SBI_CLK1_Ex4_TPM.RDS"),
                               "multivariate",
                               years_col = "OS_years",
                               status_col = "OS_status")

forest_os <- plotForest(readRDS(file.path(results_dir, "cox_OS_additive_terms_resection_lgg_group_cluster_SBI_CLK1_Ex4_TPM.RDS")))

Warning: The `guide` argument in `scale_*()` cannot be `FALSE`. This was deprecated in ggplot2 3.3.4.
ℹ Please use "none" instead.
This warning is displayed once every 8 hours.
Call `lifecycle::last_lifecycle_warnings()` to see where this warning was generated.

Warning: Removed 3 rows containing missing values or values outside the scale
range (`geom_errorbarh()`).

Warning: Removed 3 rows containing missing values or values outside the scale
range (`geom_text()`).

forest_os

ggsave(file.path(plot_dir, "forest_add_OS_resection_lgg_group_cluster_assignment_SBI_CLK1_Ex4_TPM.pdf"),
       forest_os,
       width = 10, height = 6, units = "in",
       device = "pdf")

add_model_efs <- fit_save_model(metadata[!metadata$extent_of_tumor_resection %in% c("Not Reported", "Unavailable"),],
                              terms = "extent_of_tumor_resection+lgg_group+cluster+age_at_diagnosis_years+SBI_SE+CLK1_Ex4_TPM",
                               file.path(results_dir, "cox_EFS_additive_terms_resection_lgg_group_cluster_SBI_CLK1_Ex4_TPM.RDS"),
                               "multivariate",
                               years_col = "EFS_years",
                               status_col = "EFS_status")

forest_efs <- plotForest(readRDS(file.path(results_dir, "cox_EFS_additive_terms_resection_lgg_group_cluster_SBI_CLK1_Ex4_TPM.RDS")))

Warning: Removed 3 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 3 rows containing missing values or values outside the scale range (`geom_text()`).

forest_efs

ggsave(file.path(plot_dir, "forest_add_EFS_resection_lgg_group_cluster_assignment_SBI_CLK1_Ex4_TPM.pdf"),
       forest_efs,
       width = 10, height = 6, units = "in",
       device = "pdf")

repeat analysis with CLK1 exon 4 TPM alone

add_model_os <- fit_save_model(metadata[!metadata$extent_of_tumor_resection %in% c("Not Reported", "Unavailable"),],
                              terms = "extent_of_tumor_resection+lgg_group+cluster+age_at_diagnosis_years+CLK1_Ex4_TPM",
                               file.path(results_dir, "cox_OS_additive_terms_resection_lgg_group_cluster_CLK1_Ex4_TPM.RDS"),
                               "multivariate",
                               years_col = "OS_years",
                               status_col = "OS_status")

forest_os <- plotForest(readRDS(file.path(results_dir, "cox_OS_additive_terms_resection_lgg_group_cluster_CLK1_Ex4_TPM.RDS")))

Warning: Removed 3 rows containing missing values or values outside the scale
range (`geom_errorbarh()`).

Warning: Removed 3 rows containing missing values or values outside the scale
range (`geom_text()`).

forest_os

ggsave(file.path(plot_dir, "forest_add_OS_resection_lgg_group_cluster_assignment_CLK1_Ex4_TPM.pdf"),
       forest_os,
       width = 10, height = 6, units = "in",
       device = "pdf")


add_model_efs <- fit_save_model(metadata[!metadata$extent_of_tumor_resection %in% c("Not Reported", "Unavailable"),],
                              terms = "extent_of_tumor_resection+lgg_group+cluster+age_at_diagnosis_years+CLK1_Ex4_TPM",
                               file.path(results_dir, "cox_EFS_additive_terms_resection_lgg_group_cluster_CLK1_Ex4_TPM.RDS"),
                               "multivariate",
                               years_col = "EFS_years",
                               status_col = "EFS_status")

forest_efs <- plotForest(readRDS(file.path(results_dir, "cox_EFS_additive_terms_resection_lgg_group_cluster_CLK1_Ex4_TPM.RDS")))

Warning: Removed 3 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 3 rows containing missing values or values outside the scale range (`geom_text()`).

forest_efs

ggsave(file.path(plot_dir, "forest_add_EFS_resection_lgg_group_cluster_assignment_CLK1_Ex4_TPM.pdf"),
       forest_efs,
       width = 10, height = 6, units = "in",
       device = "pdf")

repeat analysis with CLK1 exon 4 PSI

add_model_os <- fit_save_model(metadata[!metadata$extent_of_tumor_resection %in% c("Not Reported", "Unavailable"),],
                              terms = "extent_of_tumor_resection+lgg_group+cluster+age_at_diagnosis_years+CLK1_ex4_PSI",
                               file.path(results_dir, "cox_OS_additive_terms_resection_lgg_group_cluster_CLK1_ex4_PSI.RDS"),
                               "multivariate",
                               years_col = "OS_years",
                               status_col = "OS_status")

forest_os <- plotForest(readRDS(file.path(results_dir, "cox_OS_additive_terms_resection_lgg_group_cluster_CLK1_ex4_PSI.RDS")))

Warning: Removed 3 rows containing missing values or values outside the scale
range (`geom_errorbarh()`).

Warning: Removed 3 rows containing missing values or values outside the scale
range (`geom_text()`).

forest_os

ggsave(file.path(plot_dir, "forest_add_OS_resection_lgg_group_cluster_assignment_CLK1_ex4_PSI.pdf"),
       forest_os,
       width = 10, height = 6, units = "in",
       device = "pdf")

add_model_efs <- fit_save_model(metadata[!metadata$extent_of_tumor_resection %in% c("Not Reported", "Unavailable"),],
                              terms = "extent_of_tumor_resection+lgg_group+cluster+age_at_diagnosis_years+CLK1_ex4_PSI",
                               file.path(results_dir, "cox_EFS_additive_terms_resection_lgg_group_cluster_CLK1_ex4_PSI.RDS"),
                               "multivariate",
                               years_col = "EFS_years",
                               status_col = "EFS_status")

forest_efs <- plotForest(readRDS(file.path(results_dir, "cox_EFS_additive_terms_resection_lgg_group_cluster_CLK1_ex4_PSI.RDS")))

Warning: Removed 3 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 3 rows containing missing values or values outside the scale range (`geom_text()`).

forest_efs

ggsave(file.path(plot_dir, "forest_add_EFS_resection_lgg_group_cluster_assignment_CLK1_ex4_PSI.pdf"),
       forest_efs,
       width = 10, height = 6, units = "in",
       device = "pdf")

Interaction with GSVA, SBI, CLK1

models <- c("spliceosome_gsva_score", "SBI_SE", "CLK1_Ex4_TPM", "CLK1_ex4_PSI")
# by cluster
for (each in models) {
  int_model_efs <- fit_save_model(metadata[!metadata$extent_of_tumor_resection %in% c("Not Reported", "Unavailable"),],
                                terms = paste0("extent_of_tumor_resection+lgg_group+cluster*", each, "+age_at_diagnosis_years"),
                                 file.path(results_dir, paste0("cox_EFS_interaction_terms_resection_lgg_group_cluster_", each, ".RDS")),
                                 "multivariate",
                                 years_col = "EFS_years",
                                 status_col = "EFS_status")
  
  int_forest_efs <- plotForest(readRDS(file.path(results_dir, paste0("cox_EFS_interaction_terms_resection_lgg_group_cluster_", each, ".RDS"))))
  
  int_forest_efs
  
  ggsave(file.path(plot_dir, paste0("forest_int_EFS_resection_lgg_group_cluster_assignment_", each, ".pdf")),
         int_forest_efs,
         width = 10, height = 6, units = "in",
         device = "pdf")

  int_model_os <- fit_save_model(metadata[!metadata$extent_of_tumor_resection %in% c("Not Reported", "Unavailable"),],
                                terms = paste0("extent_of_tumor_resection+lgg_group+cluster*", each, "+age_at_diagnosis_years"),
                                 file.path(results_dir, paste0("cox_OS_interaction_terms_resection_lgg_group_cluster_", each, ".RDS")),
                                 "multivariate",
                                 years_col = "OS_years",
                                 status_col = "OS_status")
  
  int_forest_os <- plotForest(readRDS(file.path(results_dir, paste0("cox_OS_interaction_terms_resection_lgg_group_cluster_", each, ".RDS"))))
  
  int_forest_os
  
  ggsave(file.path(plot_dir, paste0("forest_int_OS_resection_lgg_group_cluster_assignment_", each, ".pdf")),
         int_forest_os,
         width = 10, height = 6, units = "in",
         device = "pdf")
}

Warning: Removed 3 rows containing missing values or values outside the scale
range (`geom_errorbarh()`).

Warning: Removed 3 rows containing missing values or values outside the scale
range (`geom_text()`).

Warning: Removed 3 rows containing missing values or values outside the scale
range (`geom_errorbarh()`).

Warning: Removed 3 rows containing missing values or values outside the scale
range (`geom_text()`).

Warning: Removed 3 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 3 rows containing missing values or values outside the scale range (`geom_text()`).

Warning: Removed 3 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 3 rows containing missing values or values outside the scale range (`geom_text()`).

Warning: Removed 3 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 3 rows containing missing values or values outside the scale range (`geom_text()`).

Warning: Removed 3 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 3 rows containing missing values or values outside the scale range (`geom_text()`).

Warning: Removed 3 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 3 rows containing missing values or values outside the scale range (`geom_text()`).

Warning: Removed 3 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 3 rows containing missing values or values outside the scale range (`geom_text()`).

## clk1 x age
  int_model_efs <- fit_save_model(metadata[!metadata$extent_of_tumor_resection %in% c("Not Reported", "Unavailable"),],
                                terms = paste0("extent_of_tumor_resection+lgg_group+cluster+CLK1_Ex4_TPM*age_at_diagnosis_years"),
                                 file.path(results_dir, paste0("cox_EFS_interaction_terms_resection_lgg_group_cluster_CLK1_Ex4_TPM_age.RDS")),
                                 "multivariate",
                                 years_col = "EFS_years",
                                 status_col = "EFS_status")
  
  int_forest_efs <- plotForest(readRDS(file.path(results_dir, paste0("cox_EFS_interaction_terms_resection_lgg_group_cluster_CLK1_Ex4_TPM_age.RDS"))))

Warning: Removed 3 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 3 rows containing missing values or values outside the scale range (`geom_text()`).

  int_forest_efs

  ggsave(file.path(plot_dir, paste0("forest_int_EFS_resection_lgg_group_cluster_clk1_age.pdf")),
         int_forest_efs,
         width = 10, height = 6, units = "in",
         device = "pdf")

  int_model_os <- fit_save_model(metadata[!metadata$extent_of_tumor_resection %in% c("Not Reported", "Unavailable"),],
                                terms = paste0("extent_of_tumor_resection+lgg_group+cluster+CLK1_Ex4_TPM*age_at_diagnosis_years"),
                                 file.path(results_dir, paste0("cox_OS_interaction_terms_resection_lgg_group_cluster_clk1_age.RDS")),
                                 "multivariate",
                                 years_col = "OS_years",
                                 status_col = "OS_status")
  
  int_forest_os <- plotForest(readRDS(file.path(results_dir, paste0("cox_OS_interaction_terms_resection_lgg_group_cluster_clk1_age.RDS"))))

Warning: Removed 3 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 3 rows containing missing values or values outside the scale range (`geom_text()`).

  int_forest_os

  ggsave(file.path(plot_dir, paste0("forest_int_OS_resection_lgg_group_cluster_clk1_age.pdf")),
         int_forest_os,
         width = 10, height = 6, units = "in",
         device = "pdf")

models2 <- c("SBI_SE", "CLK1_Ex4_TPM")
for (each in models2) {
  #### by spliceosome_gsva_score
  int_model_efs <- fit_save_model(metadata[!metadata$extent_of_tumor_resection %in% c("Not Reported", "Unavailable"),],
                                terms = paste0("extent_of_tumor_resection+lgg_group+cluster+spliceosome_gsva_score*", each, "+age_at_diagnosis_years"),
                                 file.path(results_dir, paste0("cox_EFS_interaction_terms_resection_lgg_group_cluster_spliceosome_gsva_score_", each, ".RDS")),
                                 "multivariate",
                                 years_col = "EFS_years",
                                 status_col = "EFS_status")
  
  int_forest_efs <- plotForest(readRDS(file.path(results_dir, paste0("cox_EFS_interaction_terms_resection_lgg_group_cluster_spliceosome_gsva_score_", each, ".RDS"))))
  
  int_forest_efs
  
  ggsave(file.path(plot_dir, paste0("cox_EFS_interaction_terms_resection_lgg_group_cluster_spliceosome_gsva_score_", each, ".pdf")),
         int_forest_efs,
         width = 10, height = 6, units = "in",
         device = "pdf")
}

Warning: Removed 3 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 3 rows containing missing values or values outside the scale range (`geom_text()`).

Warning: Removed 3 rows containing missing values or values outside the scale
range (`geom_errorbarh()`).

Warning: Removed 3 rows containing missing values or values outside the scale
range (`geom_text()`).

add_model_efs <- fit_save_model(metadata[!metadata$extent_of_tumor_resection %in% c("Not Reported", "Unavailable"),],
                              terms = "extent_of_tumor_resection+lgg_group+cluster+age_at_diagnosis_years+spliceosome_gsva_score+CLK1_Ex4_TPM",
                               file.path(results_dir, "cox_EFS_additive_terms_resection_lgg_group_cluster_spliceosome_score_CLK1_Ex4_TPM.RDS"),
                               "multivariate",
                               years_col = "EFS_years",
                               status_col = "EFS_status")

forest_efs <- plotForest(readRDS(file.path(results_dir, "cox_EFS_additive_terms_resection_lgg_group_cluster_spliceosome_score_CLK1_Ex4_TPM.RDS")))

Warning: Removed 3 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 3 rows containing missing values or values outside the scale range (`geom_text()`).

forest_efs

ggsave(file.path(plot_dir, "forest_add_EFS_resection_lgg_group_cluster_assignment_spliceosome_score_CLK1_Ex4_TPM.pdf"),
       forest_efs,
       width = 10, height = 6, units = "in",
       device = "pdf")


add_model_os <- fit_save_model(metadata[!metadata$extent_of_tumor_resection %in% c("Not Reported", "Unavailable"),],
                              terms = "extent_of_tumor_resection+lgg_group+cluster+age_at_diagnosis_years+spliceosome_gsva_score+CLK1_Ex4_TPM",
                               file.path(results_dir, "cox_OS_additive_terms_resection_lgg_group_cluster_spliceosome_score_CLK1_Ex4_TPM.RDS"),
                               "multivariate",
                               years_col = "EFS_years",
                               status_col = "EFS_status")

forest_os <- plotForest(readRDS(file.path(results_dir, "cox_OS_additive_terms_resection_lgg_group_cluster_spliceosome_score_CLK1_Ex4_TPM.RDS")))

Warning: Removed 3 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 3 rows containing missing values or values outside the scale range (`geom_text()`).

forest_os

ggsave(file.path(plot_dir, "forest_add_OS_resection_lgg_group_cluster_assignment_spliceosome_score_CLK1_Ex4_TPM.pdf"),
       forest_efs,
       width = 10, height = 6, units = "in",
       device = "pdf")

Filter for cluster 6

cluster6_df <- metadata %>%
  dplyr::filter(cluster == "Cluster 6",
                !is.na(EFS_days)) %>%
  dplyr::mutate(CLK1_TPM_group = case_when(
      CLK1_Ex4_TPM > summary(CLK1_Ex4_TPM)["3rd Qu."] ~ "High CLK1 TPM",
      CLK1_Ex4_TPM < summary(CLK1_Ex4_TPM)["1st Qu."] ~ "Low CLK1 TPM",
      TRUE ~ NA_character_),
      CLK1_PSI_group = case_when(CLK1_ex4_PSI > summary(CLK1_ex4_PSI)["3rd Qu."] ~ "High CLK1 PSI",
      CLK1_ex4_PSI < summary(CLK1_ex4_PSI)["1st Qu."] ~ "Low CLK1 PSI",
      TRUE ~ NA_character_
    )) %>%
  dplyr::mutate(CLK1_TPM_group = fct_relevel(CLK1_TPM_group,
                                                 c("Low CLK1 TPM", "High CLK1 TPM")),
                CLK1_PSI_group = fct_relevel(CLK1_PSI_group,
                                                 c("Low CLK1 PSI", "High CLK1 PSI")))

Generate KM models with CLK1_TPM_group as covariate

# Generate kaplan meier survival models for OS and EFS, and save outputs
c6_clk_tpm_kap_os <- survival_analysis(
  metadata  = cluster6_df %>% 
    dplyr::filter(!is.na(CLK1_TPM_group)),
  ind_var = "CLK1_TPM_group",
  test = "kap.meier",
  metadata_sample_col = "Kids_First_Biospecimen_ID",
  days_col = "OS_days",
  status_col = "OS_status"
)

Warning: Using an external vector in selections was deprecated in tidyselect 1.1.0.
ℹ Please use `all_of()` or `any_of()` instead.
  # Was:
  data %>% select(ind_var)

  # Now:
  data %>% select(all_of(ind_var))

See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
This warning is displayed once every 8 hours.
Call `lifecycle::last_lifecycle_warnings()` to see where this warning was generated.

Testing model: survival::Surv(OS_days, OS_status) ~ CLK1_TPM_group with kap.meier

readr::write_rds(c6_clk_tpm_kap_os,
                 file.path(results_dir, "logrank_cluster6_OS_clk1_tpm_group.RDS"))

c6_clk_tpm_kap_efs <- survival_analysis(
  metadata  = cluster6_df %>% 
    dplyr::filter(!is.na(CLK1_TPM_group)),
  ind_var = "CLK1_TPM_group",
  test = "kap.meier",
  metadata_sample_col = "Kids_First_Biospecimen_ID",
  days_col = "EFS_days",
  status_col = "EFS_status"
)

Testing model: survival::Surv(EFS_days, EFS_status) ~ CLK1_TPM_group with kap.meier

readr::write_rds(c6_clk_tpm_kap_efs,
                 file.path(results_dir, "logrank_cluster6_EFS_clk1_tpm_group.RDS"))

Generate KM models with CLK1_PSI_group as covariate

# Generate kaplan meier survival models for OS and EFS, and save outputs
c6_clk_psi_kap_os <- survival_analysis(
  metadata  = cluster6_df %>% 
    dplyr::filter(!is.na(CLK1_PSI_group)),
  ind_var = "CLK1_PSI_group",
  test = "kap.meier",
  metadata_sample_col = "Kids_First_Biospecimen_ID",
  days_col = "OS_days",
  status_col = "OS_status"
)

Testing model: survival::Surv(OS_days, OS_status) ~ CLK1_PSI_group with kap.meier

readr::write_rds(c6_clk_psi_kap_os,
                 file.path(results_dir, "logrank_cluster6_OS_clk1_psi_group.RDS"))

c6_clk_psi_kap_efs <- survival_analysis(
  metadata  = cluster6_df %>% 
    dplyr::filter(!is.na(CLK1_PSI_group)),
  ind_var = "CLK1_PSI_group",
  test = "kap.meier",
  metadata_sample_col = "Kids_First_Biospecimen_ID",
  days_col = "EFS_days",
  status_col = "EFS_status"
)

Testing model: survival::Surv(EFS_days, EFS_status) ~ CLK1_PSI_group with kap.meier

readr::write_rds(c6_clk_psi_kap_efs,
                 file.path(results_dir, "logrank_cluster6_EFS_clk1_psi_group.RDS"))

Generate Cluster 6 KM SI_group plots

km_c6_clk_tpm_os_plot <- plotKM(model = c6_clk_tpm_kap_os,
                    variable = "CLK1_TPM_group",
                    combined = F, 
                    title = "Cluster 6, overall survival",
                    p_pos = "topright")

Warning: No shared levels found between `names(values)` of the manual scale and
the data's fill values.

ggsave(file.path(plot_dir, "km_cluster6_OS_clk1_tpm_group.pdf"),
       km_c6_clk_tpm_os_plot,
       width = 8, height = 5, units = "in",
       device = "pdf")

Warning: No shared levels found between `names(values)` of the manual scale and
the data's fill values.

Warning: No shared levels found between `names(values)` of the manual scale and
the data's colour values.

Warning: No shared levels found between `names(values)` of the manual scale and
the data's fill values.

km_c6_clk1_tpm_efs_plot <- plotKM(model = c6_clk_tpm_kap_efs,
                    variable = "CLK1_TPM_group",
                    combined = F, 
                    title = "Cluster 6, event-free survival",
                    p_pos = "topright")

Warning: No shared levels found between `names(values)` of the manual scale and
the data's fill values.

ggsave(file.path(plot_dir, "km_cluster6_EFS_clk1_tpm_group.pdf"), 
       km_c6_clk1_tpm_efs_plot,
       width = 8, height = 5, units = "in",
       device = "pdf")

Warning: No shared levels found between `names(values)` of the manual scale and
the data's fill values.

Warning: No shared levels found between `names(values)` of the manual scale and
the data's colour values.

Warning: No shared levels found between `names(values)` of the manual scale and
the data's fill values.

km_c6_clk1_psi_os_plot <- plotKM(model = c6_clk_psi_kap_os,
                    variable = "CLK1_PSI_group",
                    combined = F, 
                    title = "Cluster 6, overall survival",
                    p_pos = "topright")

Warning: No shared levels found between `names(values)` of the manual scale and
the data's fill values.

ggsave(file.path(plot_dir, "km_cluster6_OS_clk1_psi_group.pdf"),
       km_c6_clk1_psi_os_plot,
       width = 8, height = 5, units = "in",
       device = "pdf")

Warning: No shared levels found between `names(values)` of the manual scale and
the data's fill values.

Warning: No shared levels found between `names(values)` of the manual scale and
the data's colour values.

Warning: No shared levels found between `names(values)` of the manual scale and
the data's fill values.

km_c6_clk1_psi_efs_plot <- plotKM(model = c6_clk_psi_kap_efs,
                    variable = "CLK1_PSI_group",
                    combined = F, 
                    title = "Cluster 6, event-free survival",
                    p_pos = "topright")

Warning: No shared levels found between `names(values)` of the manual scale and
the data's fill values.

ggsave(file.path(plot_dir, "km_cluster6_EFS_clk1_psi_group.pdf"), 
       km_c6_clk1_psi_efs_plot,
       width = 8, height = 5, units = "in",
       device = "pdf")

Warning: No shared levels found between `names(values)` of the manual scale and
the data's fill values.

Warning: No shared levels found between `names(values)` of the manual scale and
the data's colour values.

Warning: No shared levels found between `names(values)` of the manual scale and
the data's fill values.

Assess EFS and OS by CLK1 TPM group in multivariate models and generate forest plots

add_model_c6_efs <- fit_save_model(cluster6_df %>% 
                                      dplyr::filter(extent_of_tumor_resection != "Unavailable",
                                                    CLK1_TPM_group %in% c("High CLK1 TPM", "Low CLK1 TPM")) %>%
                                      dplyr::mutate(Histology = fct_relevel(Histology,
                                                                            c("Other high-grade glioma", "Atypical Teratoid Rhabdoid Tumor",
                                                                              "DIPG or DMG", "Ependymoma", "Mesenchymal tumor",
                                                                              "Other CNS embryonal tumor", "Low-grade glioma"))),
                              terms = "extent_of_tumor_resection+age_at_diagnosis_years+Histology+CLK1_TPM_group",
                               file.path(results_dir, "cox_EFS_additive_terms_subtype_cluster_clk1_tpm_group.RDS"),
                               "multivariate",
                               years_col = "EFS_years",
                               status_col = "EFS_status")

Warning: There was 1 warning in `dplyr::mutate()`.
ℹ In argument: `Histology = fct_relevel(...)`.
Caused by warning:
! 2 unknown levels in `f`: Mesenchymal tumor and Low-grade glioma

forest_c6_clk1_efs <- plotForest(readRDS(file.path(results_dir, "cox_EFS_additive_terms_subtype_cluster_clk1_tpm_group.RDS")))

Warning: Removed 3 rows containing missing values or values outside the scale
range (`geom_errorbarh()`).

Warning: Removed 3 rows containing missing values or values outside the scale
range (`geom_text()`).

ggsave(file.path(plot_dir, "forest_add_EFS_cluster6_histology_resection_clk1_tpm_group.pdf"),
       forest_c6_clk1_efs,
       width = 9, height = 4, units = "in",
       device = "pdf")

add_model_c6_os <- fit_save_model(cluster6_df %>% 
                                    dplyr::filter(!extent_of_tumor_resection %in% c("Not Reported", "Unavailable")) %>%
                                    dplyr::mutate(Histology = fct_relevel(Histology,
                                                                            c("Other high-grade glioma", "Atypical Teratoid Rhabdoid Tumor",
                                                                              "DIPG or DMG", "Ependymoma", "Mesenchymal tumor",
                                                                              "Other CNS embryonal tumor", "Low-grade glioma"))),
                              terms = "extent_of_tumor_resection+age_at_diagnosis_years+Histology+CLK1_TPM_group",
                               file.path(results_dir, "cox_OS_additive_terms_subtype_cluster_clk1_tpm_group.RDS"),
                               "multivariate",
                               years_col = "OS_years",
                               status_col = "OS_status")

forest_c6_clk1_os <- plotForest(readRDS(file.path(results_dir, "cox_OS_additive_terms_subtype_cluster_clk1_tpm_group.RDS")))

Warning: Removed 3 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 3 rows containing missing values or values outside the scale range (`geom_text()`).

ggsave(file.path(plot_dir, "forest_add_OS_cluster6_histology_resection_clk1_tpm_group.pdf"),
       forest_c6_clk1_os,
       width = 9, height = 4, units = "in",
       device = "pdf")

Assess EFS and OS by CLK1 PSI group in multivariate models and generate forest plots

add_model_c6_efs <- fit_save_model(cluster6_df %>% 
                                      dplyr::filter(extent_of_tumor_resection != "Unavailable",
                                                    CLK1_PSI_group %in% c("High CLK1 PSI", "Low CLK1 PSI")) %>%
                                      dplyr::mutate(Histology = fct_relevel(Histology,
                                                                            c("Other high-grade glioma", "Atypical Teratoid Rhabdoid Tumor",
                                                                              "DIPG or DMG", "Ependymoma", "Mesenchymal tumor",
                                                                              "Other CNS embryonal tumor", "Low-grade glioma"))),
                              terms = "extent_of_tumor_resection+age_at_diagnosis_years+Histology+CLK1_PSI_group",
                               file.path(results_dir, "cox_EFS_additive_terms_subtype_cluster_clk1_psi_group.RDS"),
                               "multivariate",
                               years_col = "EFS_years",
                               status_col = "EFS_status")

Warning: There was 1 warning in `dplyr::mutate()`.
ℹ In argument: `Histology = fct_relevel(...)`.
Caused by warning:
! 2 unknown levels in `f`: Mesenchymal tumor and Low-grade glioma

Warning in coxph.fit(X, Y, istrat, offset, init, control, weights = weights, :
Loglik converged before variable 6 ; coefficient may be infinite.

forest_c6_clk1_efs <- plotForest(readRDS(file.path(results_dir, "cox_EFS_additive_terms_subtype_cluster_clk1_psi_group.RDS")))

Warning: Removed 3 rows containing missing values or values outside the scale
range (`geom_errorbarh()`).

Warning: Removed 3 rows containing missing values or values outside the scale
range (`geom_text()`).

ggsave(file.path(plot_dir, "forest_add_EFS_cluster6_histology_resection_clk1_psi_group.pdf"),
       forest_c6_clk1_efs,
       width = 9, height = 4, units = "in",
       device = "pdf")

add_model_c6_os <- fit_save_model(cluster6_df %>% 
                                    dplyr::filter(!extent_of_tumor_resection %in% c("Not Reported", "Unavailable")) %>%
                                    dplyr::mutate(Histology = fct_relevel(Histology,
                                                                            c("Other high-grade glioma", "Atypical Teratoid Rhabdoid Tumor",
                                                                              "DIPG or DMG", "Ependymoma", "Mesenchymal tumor",
                                                                              "Other CNS embryonal tumor", "Low-grade glioma"))),
                              terms = "extent_of_tumor_resection+age_at_diagnosis_years+Histology+CLK1_PSI_group",
                               file.path(results_dir, "cox_OS_additive_terms_subtype_cluster_clk1_psi_group.RDS"),
                               "multivariate",
                               years_col = "OS_years",
                               status_col = "OS_status")

Warning in coxph.fit(X, Y, istrat, offset, init, control, weights = weights, :
Loglik converged before variable 6 ; coefficient may be infinite.

forest_c6_clk1_os <- plotForest(readRDS(file.path(results_dir, "cox_OS_additive_terms_subtype_cluster_clk1_psi_group.RDS")))

Warning: Removed 3 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 3 rows containing missing values or values outside the scale range (`geom_text()`).

ggsave(file.path(plot_dir, "forest_add_OS_cluster6_histology_resection_clk1_psi_group.pdf"),
       forest_c6_clk1_os,
       width = 9, height = 4, units = "in",
       device = "pdf")

Assess EFS and OS by CLK1 ex 4 TPM in multivariate models and generate forest plots

add_model_c6_efs <- fit_save_model(cluster6_df %>% 
                                      dplyr::filter(extent_of_tumor_resection != "Unavailable") %>%
                                      dplyr::mutate(Histology = fct_relevel(Histology,
                                                                            c("Other high-grade glioma", "Atypical Teratoid Rhabdoid Tumor",
                                                                              "DIPG or DMG", "Ependymoma", "Mesenchymal tumor",
                                                                              "Other CNS embryonal tumor", "Low-grade glioma"))),
                              terms = "extent_of_tumor_resection+age_at_diagnosis_years+Histology+CLK1_Ex4_TPM",
                               file.path(results_dir, "cox_EFS_additive_terms_subtype_cluster_clk1_tpm.RDS"),
                               "multivariate",
                               years_col = "EFS_years",
                               status_col = "EFS_status")

Warning in coxph.fit(X, Y, istrat, offset, init, control, weights = weights, :
Loglik converged before variable 9 ; coefficient may be infinite.

forest_c6_clk1_efs <- plotForest(readRDS(file.path(results_dir, "cox_EFS_additive_terms_subtype_cluster_clk1_tpm.RDS")))

Warning: Removed 2 rows containing missing values or values outside the scale
range (`geom_errorbarh()`).

Warning: Removed 2 rows containing missing values or values outside the scale
range (`geom_text()`).

ggsave(file.path(plot_dir, "forest_add_EFS_cluster6_histology_resection_clk1_tpm.pdf"),
       forest_c6_clk1_efs,
       width = 9, height = 4, units = "in",
       device = "pdf")

add_model_c6_os <- fit_save_model(cluster6_df %>% 
                                    dplyr::filter(!extent_of_tumor_resection %in% c("Not Reported", "Unavailable")) %>%
                                    dplyr::mutate(Histology = fct_relevel(Histology,
                                                                            c("Other high-grade glioma", "Atypical Teratoid Rhabdoid Tumor",
                                                                              "DIPG or DMG", "Ependymoma", "Mesenchymal tumor",
                                                                              "Other CNS embryonal tumor", "Low-grade glioma"))),
                              terms = "extent_of_tumor_resection+age_at_diagnosis_years+Histology+CLK1_Ex4_TPM",
                               file.path(results_dir, "cox_OS_additive_terms_subtype_cluster_clk1_tpm.RDS"),
                               "multivariate",
                               years_col = "OS_years",
                               status_col = "OS_status")

Warning in coxph.fit(X, Y, istrat, offset, init, control, weights = weights, :
Loglik converged before variable 9 ; coefficient may be infinite.

forest_c6_clk1_os <- plotForest(readRDS(file.path(results_dir, "cox_OS_additive_terms_subtype_cluster_clk1_tpm.RDS")))

Warning: Removed 2 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 2 rows containing missing values or values outside the scale range (`geom_text()`).

ggsave(file.path(plot_dir, "forest_add_OS_cluster6_histology_resection_clk1_tpm.pdf"),
       forest_c6_clk1_os,
       width = 9, height = 4, units = "in",
       device = "pdf")

Assess EFS and OS by CLK1 ex 4 PSI in multivariate models and generate forest plots

add_model_c6_efs <- fit_save_model(cluster6_df %>% 
                                      dplyr::filter(extent_of_tumor_resection != "Unavailable") %>%
                                      dplyr::mutate(Histology = fct_relevel(Histology,
                                                                            c("Other high-grade glioma", "Atypical Teratoid Rhabdoid Tumor",
                                                                              "DIPG or DMG", "Ependymoma", "Mesenchymal tumor",
                                                                              "Other CNS embryonal tumor", "Low-grade glioma"))),
                              terms = "extent_of_tumor_resection+age_at_diagnosis_years+Histology+CLK1_ex4_PSI",
                               file.path(results_dir, "cox_EFS_additive_terms_subtype_cluster_clk1_psi.RDS"),
                               "multivariate",
                               years_col = "EFS_years",
                               status_col = "EFS_status")

Warning in coxph.fit(X, Y, istrat, offset, init, control, weights = weights, :
Loglik converged before variable 9 ; coefficient may be infinite.

forest_c6_clk1_efs <- plotForest(readRDS(file.path(results_dir, "cox_EFS_additive_terms_subtype_cluster_clk1_psi.RDS")))

Warning: Removed 2 rows containing missing values or values outside the scale
range (`geom_errorbarh()`).

Warning: Removed 2 rows containing missing values or values outside the scale
range (`geom_text()`).

ggsave(file.path(plot_dir, "forest_add_EFS_cluster6_histology_resection_clk1_psi.pdf"),
       forest_c6_clk1_efs,
       width = 9, height = 4, units = "in",
       device = "pdf")

add_model_c6_os <- fit_save_model(cluster6_df %>% 
                                    dplyr::filter(!extent_of_tumor_resection %in% c("Not Reported", "Unavailable")) %>%
                                    dplyr::mutate(Histology = fct_relevel(Histology,
                                                                            c("Other high-grade glioma", "Atypical Teratoid Rhabdoid Tumor",
                                                                              "DIPG or DMG", "Ependymoma", "Mesenchymal tumor",
                                                                              "Other CNS embryonal tumor", "Low-grade glioma"))),
                              terms = "extent_of_tumor_resection+age_at_diagnosis_years+Histology+CLK1_ex4_PSI",
                               file.path(results_dir, "cox_OS_additive_terms_subtype_cluster_clk1_psi.RDS"),
                               "multivariate",
                               years_col = "OS_years",
                               status_col = "OS_status")

Warning in coxph.fit(X, Y, istrat, offset, init, control, weights = weights, :
Loglik converged before variable 9 ; coefficient may be infinite.

forest_c6_clk1_os <- plotForest(readRDS(file.path(results_dir, "cox_OS_additive_terms_subtype_cluster_clk1_psi.RDS")))

Warning: Removed 2 rows containing missing values or values outside the scale range (`geom_errorbarh()`).
Removed 2 rows containing missing values or values outside the scale range (`geom_text()`).

ggsave(file.path(plot_dir, "forest_add_OS_cluster6_histology_resection_clk1_psi.pdf"),
       forest_c6_clk1_os,
       width = 9, height = 4, units = "in",
       device = "pdf")

Print session info

sessionInfo()

R version 4.4.0 (2024-04-24)
Platform: x86_64-pc-linux-gnu
Running under: Ubuntu 22.04.4 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3 
LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so;  LAPACK version 3.10.0

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

time zone: Etc/UTC
tzcode source: system (glibc)

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] gtools_3.9.5    survminer_0.4.9 patchwork_1.2.0 ggpubr_0.6.0   
 [5] survival_3.7-0  lubridate_1.9.4 forcats_1.0.0   stringr_1.5.1  
 [9] dplyr_1.1.4     purrr_1.0.4     readr_2.1.5     tidyr_1.3.1    
[13] tibble_3.2.1    ggplot2_3.5.1   tidyverse_2.0.0

loaded via a namespace (and not attached):
 [1] gtable_0.3.6      xfun_0.50         bslib_0.9.0       rstatix_0.7.2    
 [5] lattice_0.22-6    tzdb_0.4.0        vctrs_0.6.5       tools_4.4.0      
 [9] generics_0.1.3    parallel_4.4.0    pkgconfig_2.0.3   Matrix_1.7-2     
[13] data.table_1.16.4 lifecycle_1.0.4   farver_2.1.2      compiler_4.4.0   
[17] textshaping_1.0.0 munsell_0.5.1     carData_3.0-5     colorblindr_0.1.0
[21] htmltools_0.5.8.1 sass_0.4.9        yaml_2.3.10       crayon_1.5.3     
[25] pillar_1.10.1     car_3.1-2         jquerylib_0.1.4   cachem_1.1.0     
[29] abind_1.4-5       km.ci_0.5-6       commonmark_1.9.2  tidyselect_1.2.1 
[33] digest_0.6.37     stringi_1.8.4     labeling_0.4.3    splines_4.4.0    
[37] cowplot_1.1.3     rprojroot_2.0.4   fastmap_1.2.0     grid_4.4.0       
[41] colorspace_2.1-1  cli_3.6.4         magrittr_2.0.3    broom_1.0.7      
[45] withr_3.0.2       scales_1.3.0      backports_1.5.0   bit64_4.6.0-1    
[49] timechange_0.3.0  rmarkdown_2.29    ggtext_0.1.2      bit_4.5.0.1      
[53] gridExtra_2.3     ggsignif_0.6.4    ragg_1.3.3        zoo_1.8-12       
[57] hms_1.1.3         evaluate_1.0.3    knitr_1.49        KMsurv_0.1-5     
[61] markdown_1.13     survMisc_0.5.6    rlang_1.1.5       Rcpp_1.0.14      
[65] gridtext_0.1.5    xtable_1.8-4      glue_1.8.0        xml2_1.3.6       
[69] vroom_1.6.5       jsonlite_1.8.9    R6_2.6.1          systemfonts_1.2.1